Effectiveness of Deep Learning Trained on SynthCity Data for Urban Point-Cloud Classification
- Oak Ridge Institute for Science and Education, Belcamp
- National Geospatial-Intelligence Agency, Saint Louis
3D object recognition is one of the most popular areas of study in computer vision. Many of the more recent algorithms focus on indoor point clouds, classifying 3D geometric objects, and segmenting outdoor 3D scenes. One of the challenges of the classification pipeline is finding adequate and accurate training data. Hence, this article seeks to evaluate the accuracy of a synthetically generated data set called SynthCity, tested on two mobile laser-scan data sets. Varying levels of noise were applied to the training data to reflect varying levels of noise in different scanners. The chosen deep-learning algorithm was Kernel Point Convolution, a convolutional neural network that uses kernel points in Euclidean space for convolution weights.
- Research Organization:
- Oak Ridge Inst. for Science and Education (ORISE), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- DOE Contract Number:
- SC0014664
- OSTI ID:
- 1983059
- Journal Information:
- Photogrammetric Engineering and Remote Sensing, Vol. 88, Issue 2; ISSN 0099-1112
- Publisher:
- Ingenta
- Country of Publication:
- United States
- Language:
- English
Similar Records
Integrate Light-Weight Deep Learning Tools with Internet of Things
Image feature extraction and galaxy classification: a novel and efficient approach with automated machine learning